研究人员推出 Motus2，这是一种新颖的、自进化的通用世界模型，专为灵巧操作任务而设计。该模型通过耦合策略、模拟器和评估器，将感知、预测、行动、评估和改进整合到一个统一的系统中。Motus2 通过模型和数据扩展来推进世界建模，并利用专家演示和次优交互进行学习。该系统在仿生平台上实现，并包含触觉反馈以增强控制。 AI
影响 这项研究推动了能够执行复杂操作任务的具身智能体的开发，可能对机器人技术和人工智能驱动的自动化产生影响。
排序理由 该集群描述了一篇详细介绍灵巧操作新模型的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2608.30237v1 Announce Type: cross Abstract: General embodied agents should perceive, predict, act, evaluate, and improve within a unified system. World models have shown great promise in building such agents, yet existing models typically append an action output head to a w…
General embodied agents should perceive, predict, act, evaluate, and improve within a unified system. World models have shown great promise in building such agents, yet existing models typically append an action output head to a world simulator, without coupling them into a close…